diff --git a/CHANGELOG.md b/CHANGELOG.md
index 6967dc814..e3b27c838 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -30,6 +30,8 @@
 
 - **Expanded warning coverage for `Sample.from_frame()` ID inference**
   - Added assertions that validate all three expected warnings are emitted when inferring an `id` column and default weights, including ID guessing, ID string casting, and automatic weight creation.
+- **Added focused unit coverage for IPW helpers**
+  - Added tests for `link_transform()`, and `calc_dev()` to validate behavior for extreme probabilities, and finite 10-fold deviance summaries.
 
 # 0.16.0 (2026-02-09)
 
diff --git a/balance/weighting_methods/ipw.py b/balance/weighting_methods/ipw.py
index 0a6a15733..a9c5377b6 100644
--- a/balance/weighting_methods/ipw.py
+++ b/balance/weighting_methods/ipw.py
@@ -30,7 +30,6 @@
 logger: logging.Logger = logging.getLogger(__package__)
 
 
-# TODO: Add tests for model_coefs()
 # TODO: Improve interpretability of model coefficients, as variables are no longer zero-centered.
 def model_coefs(
     model: ClassifierMixin,
@@ -94,7 +93,6 @@ def model_coefs(
     }
 
 
-# TODO: Add tests for link_transform()
 def link_transform(pred: np.ndarray) -> np.ndarray:
     """Transforms probabilities into log odds (link function).
 
@@ -184,7 +182,6 @@ def _convert_to_dense_array(
     return X_matrix
 
 
-# TODO: Add tests for calc_dev()
 def calc_dev(
     X_matrix: csr_matrix,
     y: np.ndarray,
diff --git a/tests/test_ipw.py b/tests/test_ipw.py
index 7c6a7004b..b8617cc0e 100644
--- a/tests/test_ipw.py
+++ b/tests/test_ipw.py
@@ -22,6 +22,7 @@
 from balance.sample_class import Sample
 from balance.weighting_methods import ipw as balance_ipw
 from packaging.version import Version
+from scipy.sparse import csr_matrix
 from sklearn.ensemble import HistGradientBoostingClassifier, RandomForestClassifier
 from sklearn.linear_model import LogisticRegression
 from sklearn.metrics import log_loss
@@ -37,6 +38,38 @@ class TestIPW(
 ):
     """Test suite for Inverse Propensity Weighting (IPW) functionality."""
 
+    def test_link_transform_handles_midpoint_and_extremes(self) -> None:
+        """link_transform should return finite log-odds for probabilities in [0, 1]."""
+
+        transformed = balance_ipw.link_transform(np.array([0.5, 0.0, 1.0]))
+        self.assertAlmostEqual(transformed[0], 0.0, places=10)
+        self.assertTrue(np.isfinite(transformed[1]))
+        self.assertTrue(np.isfinite(transformed[2]))
+        self.assertLess(transformed[1], 0)
+        self.assertGreater(transformed[2], 0)
+
+    def test_calc_dev_returns_finite_mean_and_sd(self) -> None:
+        """calc_dev should run 10-fold CV and return finite deviance summary."""
+
+        rng = np.random.RandomState(42)
+        X = rng.normal(size=(40, 2))
+        y = np.array([0] * 20 + [1] * 20)
+        foldids = np.tile(np.arange(10), 4)
+        model_weights = np.ones(40)
+
+        dev_mean, dev_sd = balance_ipw.calc_dev(
+            csr_matrix(X),
+            y,
+            LogisticRegression(random_state=0, max_iter=300),
+            model_weights,
+            foldids,
+        )
+
+        self.assertTrue(np.isfinite(dev_mean))
+        self.assertTrue(np.isfinite(dev_sd))
+        self.assertGreaterEqual(dev_mean, 0.0)
+        self.assertGreaterEqual(dev_sd, 0.0)
+
     def test_ipw_weights_order(self) -> None:
         """Test that IPW assigns correct relative weight ordering.